摘要
为解决已有PCA人脸识别算法对人脸偏转和光线变化不具有鲁棒性以及易丢失人脸结构信息的不足,提出一种基于2DGabor的改进人脸识别算法。利用2DGabor提取人脸初始特征,生成Gabor图像;利用旋转不变一致LBP算子对Gabor图像进行特征提取,分块统计特征向量;通过2DPCA算法降低特征维度,建立特征脸,输入分类器,进行匹配识别。在VS2013+OpenCV下,选用ORL、YaleA、YaleB和CMU-PIE人脸库为实验数据集,实验结果表明,与已有算法相比,在面部偏转和光线变化情况下,该算法缩短了匹配时间,提高了识别率。
To solve some deficiencies of existing face recognition algorithms on PCA,such as their non-robust performance in the face rotation and lighting change,and lost facial structures,an improved face recognition algorithm based on 2DGabor was proposed.2DGabor was used to extract initial facial features,and Gabor images were generated.The features of Gabor images were extracted using the rotation invariant uniform LBP operator,and eigenvectors were segmented into blocks.The 2DPCA algorithm was used to reduce the facial feature dimension,to establish the feature face,and to input the classifier for matching face recognition.Experiments were carried out on ORL,YaleA,YaleB and CMU-PIE database using OpenCV in VS2013.Experimental results show that the proposed algorithm can improve the recognition rate and shorten the matching time facing the face rotation and lighting change compared with existing face recognition algorithms.
引文
[1]Kamaruzaman F,Shafie A A.Recognizing faces with normalized local gabor features and spiking neuron patterns[J].Pattern Recognition,2016,53(C):102-115.
[2]LI Bin.Face recognition based on Gabor wavelet and LBP[D].Shanghai:Shanghai Normal University,2016(in Chinese).[李斌.基于Gabor小波与局部二值编码技术的人脸识别研究[D].上海:上海师范大学,2016.]
[3]YANG Sai,ZHAO Chunxia,LIU Fan,et al.A face recognition algorithm using fusion of multiple features[J].J ComputAid Desig Comput Graph,2017,29(9):1667-1672(in Chinese).[杨赛,赵春霞,刘凡,等.一种基于多种特征融合的人脸识别算法[J].计算机辅助设计与图形学学报,2017,29(9):1667-1672.]
[4]ZHOU Xi,CAO Lin.The sketch face recognition combining with Ada Boost and blocking LBP[J].Journal of Image and Graphics,2015,20(1):50-58(in Chinese).[周汐,曹林.分块LBP的素描人脸识别[J].中国图象图形学报,2015,20(1):50-58.]
[5]Ameur B,Masmoudi S,Derbel A G,et al.Fusing Gabor and LBP feature sets for KNN and SRC-based face recognition[C]//International Conference on Advanced Technologies for Signal and Image Processing.Monastir:IEEE,2016:453-458.
[6]Li C,Wei W,Li J,et al.A cloud-based monitoring system via face recognition using Gabor and CS-LBP features[J].Journal of Supercomputing,2017,73(4):1532-1546.
[7]Roy H,Bhattacharjee D.Face sketch-photo matching using the local gradient fuzzy pattern[J].IEEE Intelligent Systems,2016,31(3):30-39.
[8]WEI Li,JIANG Jianguo,QI Meibin.LDP face recognition algorithm based on polarization encoding[J].Journal of Image and Graphics,2016,21(6):756-763(in Chinese).[魏莉,蒋建国,齐美彬.偏振编码方式的LDP人脸识别算法[J].中国图象图形学报,2016,21(6):756-763.]
[9]LI Bing,SU Tianyu,YOU Lei,et al.Anti-spoofing algorithm for face using haar wavelet decomposition and uniform local binary pattern[J].Acta Scientiarum Naturalium Universitatis Nankaiensis,2018,51(3):37-43(in Chinese).[李冰,苏天宇,由磊,等.基于哈尔小波分解和等价局部2值模式的人脸防欺骗算法[J].南开大学学报(自然科学版),2018,51(3):37-43.]
[10]WANG Yan,LI Xin.Face recognition based on LDP feature and Bayesian model[J].Computer Science,2017,44(12):283-286(in Chinese).[王燕,李鑫.基于LDP特征和贝叶斯模型的人脸识别[J].计算机科学,2017,44(12):283-286.]
[11]XU Mengke,XU Daoyun,WEI Mingjun.Face recognition with Laplacian eigenmaps based on 2D-KPCA[J].Application Research of Computers,2017,34(7):2212-2215(in Chinese).[徐梦珂,许道云,魏明俊.基于2D-KPCA的拉普拉斯特征映射人脸识别[J].计算机应用研究,2017,34(7):2212-2215.]
[12]HE Yun,WU Huaiyu,ZHONG Rui.Face recognition based on ensemble learning with multiple LBP feature[J].Application Research of Computers,2018,35(1):292-295(in Chinese).[何云,吴怀宇,钟锐.基于多种LBP特征集成学习的人脸识别[J].计算机应用研究,2018,35(1):292-295.]
[13]YANG Qian,LIU Hongying.Sparse principal component analysis via regularized rank-k matrix approximation[J].Journal of Beijing University of Aeronautics and Astronautics,2017,43(6):1239-1246(in Chinese).[杨茜,刘红英.基于正则化秩k矩阵逼近的稀疏主成分分析[J].北京航空航天大学学报,2017,43(6):1239-1246.]